Gain noise in transistor based circuits

[Alan W]

Two pedals that have been getting almost non-stop use lately are the Champ Stamp and the Equilux. I can’t say enough good things about the way they sound and feel. But both share a level of noise, not dissimilar to preamp tube noise, as the level control is upped. It’s not keeping me from using them. But I’d love to get them a bit quieter. I can’t help but feel that trying out different transistors in them (thinking same type, just different transistors) would be worthwhile.

I have the DCA-75 tester. Is there a specific parameter to look for in assessing what the s/n for a given transistor will be? Would that specific parameter change depending on the circuit topology? I could blindly just start subbing in different ones (and rebiasing in the case of the Champ Stamp), but know that will be a crazy inducing task, and would love to streamline it a bit. Plus, you know, this might help me understand these circuits a bit better.

Than Champ Stamp is something else! Can’t wait to try the focus fuzz…

 

[Feral Feline]

I'm NOT the one to answer this, but I'll chuck in what little I have in my notes anyway.

CHAMP STAMP
Q1
The BC547B in the front end of the circuit has a 110 to 800 hFE according to its datasheet.
Both the BC549 and BC550 share the same gain range and are classified as "Low Noise".

Colorsound- One Knob Fuzz - Page 2

Do you mean how to build a clone of this ?: Thats very easy, as you can see here: analogguru

www.freestompboxes.org

A BC108 IS the same as a BC109 (and a BC107). They are produced all at the same time at the same wafer with the same process.

The differentation is done (only) later by selection:
High Vce specs-chips get marked BC107, low noise specs-chips get marked BC109. A BC 108 doesn´t fullfill the high Vce specs needed to become a BC107 and it doesn´t fullfill the low-noise specs needed to become a BC109 but the chip is identical.

Q2
Next in circuit is the 2N3904; you might be able to get a 2N5089 to work here, though it is more gain IIRC. I read that the 5089 is quieter than a 5088. However, the "low-noise" BC549 might be an even better fit here as well.



Q3
J201 replacements... hmm...
I read on FSB somewhere that Russian KII303A and KII303I have a low noise as well. Alas, didn't note the URL for that assertion.
KII303A,X --> J201
KII303I' --> 2N5457,2N5458

The more common usual substitution-suggestions for the J201:
2N5457, NTE458, MPF102, J113
Couldn't tell you if any of those are lower noise, but I'd bet on the MPF102 or the J113.




EQUILUX
2N7000

No real ideas. Maybe these:​

VN2222LLRLRAG MOSFET​

BS170​

I found this list of subs: BS170, NTE 491, IRF3205, IRF540N, IRF9Z34N, IRFP250N, IRFZ44 — but nothing about low noise.​

LND150-N3

Your thread for the LND150 is the only substitution reference I could find!​

 

[Alan W]

Thanks @Feral Feline. You approached this more smartly than I was thinking; my original idea was to test out all the Q1&2 that I have in stock, and cherry pick the quietest ones, but subbing a similarly spec’ed, but listed as a low noise variant, makes a lot of sense. (Not to say that the J201 couldn’t be improved on, but so far these have been in so many pedals that are all fairly quiet, I’m also thinking that this noise is more from the early stage anyway. (Not based on true understanding…).

I’ve got both 549 and 550 in stock. I’ll pull the 547, and approx. match the hfe with one of those and see what happens. (not today, today I’m trying to complete a Cattle Driver to send to a friend, and am positively in ”all thumbs” mode today! It’s amazing how many things I’ve had to pick up off the floor this morning.)

 

[finebyfine]

I’d also experiment with where your jack wiring is placed with respect to the transistors in question. Sometimes this is as effective as using shielded wire (also worth a try) in my builds in reducing noise.

Unfortunately as far as I know, noise parameters aren’t something the DCA can measure, and are in general poorly defined in datasheets, if at all. I would try a few different transistors as specified in the build docs before going for substitutes as noise can vary a decent amount from transistor to transistor of the same part number.

There are some really good noise resources, discussion and analysis in both The Art of Electronics and Douglas Self’s Small Signal Audio design

 

[Alan W]

I’d also experiment with where your jack wiring is placed with respect to the transistors in question. Sometimes this is as effective as using shielded wire (also worth a try) in my builds in reducing noise.

Unfortunately as far as I know, noise parameters aren’t something the DCA can measure, and are in general poorly defined in datasheets, if at all. I would try a few different transistors as specified in the build docs before going for substitutes as noise can vary a decent amount from transistor to transistor of the same part number.

There are some really good noise resources, discussion and analysis in both The Art of Electronics and Douglas Self’s Small Signal Audio design

I pretty much always shield my signal wires, most of my build time has been on audio equipment, so it’s a habit.

A good answer. (Even though I’m gonna hit the data sheets to see what compatible subs might be.) I have a copy of Small Signal Audio, but am not familiar with Art of Electronics—will check it out, thanks!

 

[Feral Feline]

... and am positively in ”all thumbs” mode today! It’s amazing how many things I’ve had to pick up off the floor this morning.)

I swear, I mean really curse, as my corner solder-bench without question is cursed with more gravity than most places on earth I've been.
It's amazingly annoying how much I have to pick up off the floor for every build.

 

[szukalski]

I always find transistors to add a bit of noise into things. The cleanest sounding pedals to me are always opamp based.

 

[Alan W]

I always find transistors to add a bit of noise into things. The cleanest sounding pedals to me are always opamp based.

Yeah, I expect some noise… My chop shop is very quiet though. I’m really digging the dynamics on the 2 and 3 transistor designs; I agree the op amp based designs are more faithful, especially running Burr Brown ones at 18 volts—but I’m loving the very slightly loose, but super articulate clarity of the Champ Stamp. I’m typically running it at its “least overdriven sounding“ setting, at 18v.

Looks like a good rainy day project; I’ll start by subbing in the other 547s I have, then will play with other variants. I keep hitting my head on the low ceiling of my basic understanding of electronics though—my assumption is that the bias reference given for the 547 will be a starting point on the other transistors, rather than the goal?—that some of those “maximum collector to” voltage differences on the spec sheet may affect what the optimum bias is?